Intestinal Absorption of Peptides by Coupling to Bile Acids*

Vol. 269,No. 14,Issue of April 8,pp. 10621-10627, 1994 Printed in l7,S.A.

THEJOURNAL OF B!oLocxca CHEMISTRY 0 1994 by The American Society for Biochemistry and Molecular Biology, Inc.

Intestinal Absorption of Peptides by Coupling to Bile Acids* (Received for publication, July 23, 1993, and in revised form, December 21, 1993)

Werner KramerS, Gunther Wess, Georg Neckermann, Gerrit Schubert, JurgenFink, Frank Girbig, Ulrike Gutjahr, Simone Kowalewski, Karl-Heinz Baringhaus, Georg Boger, Alfons Enhsen, Eugen Falk,Michael Friedrich, Heiner Glombik, Axel Hoffmann, Christoph Pittius, and Matthias Urmann From the Hoechst Aktiengesellschaft, 0-65926 Frankfurt am Main, Germany

For the treatment of chronic diseases the oral route is t h e Poor intestinal absorption of peptides greatly limits their use as drugs for the treatmentof chronic diseases. most desirable method of drug administration. The medicinal Since bile acids are efficiently absorbed by an active, importance of peptides as drugs will tremendously increase in Na+-dependenttransport system in the ileum of mamthe future due to the discovery of intelligent synthesis and mals,model peptides of different chain length were screening strategies for biologically active peptides(1).A major attached to the 3-position of modified3P-(w-amino- limiting factor of using peptides as drugs is their poor oral alkoxy)-7a,l2cu-dihydroxy-5~-cholan-24-oic acid.These availability and susceptibility to enzymatic hydrolysis (2, 3). A peptide-bile acid conjugates inhibited Na+-dependent variety of membrane-bound peptidases in the brush-border [SH]taurocholateuptake into brush-border membrane membrane of the intestinal absorptive epithelial cells hydrovesicles isolated from rabbit ileum in a concentration- lyze proteins and peptides to di- and tripeptides as well as dependent manner. Furthermore, photoaffinity labeling amino acids, whichare taken upby the enterocytes via specialof the bile acid-binding proteins of M , 93,000 and 14,000, a ized transport systems (4). Thus, the intestinal mucosa is identified as the protein components of the ileal Na+highly selective filter for the absorption of peptides and many dependent bile acid transport system in rabbit ileum (Kramer,W.,Girbig, F., Gutjahr, U.,Kowalewski, S., Jou- strategies to improve their intestinal absorption including coatvenal, K., Muller, G., Tripier, D., and Wess, G. (1993) J. ing with polymers (51, usage of penetration enhancers (61, or BioZ. Chem. 268,18035-18046) by the photoreactive tau- co-administration with protease inhibitors( 7 , 8 ) have been applied. A more specific way to increasethe intestinal permeabilrocholate analogue, (3,3-azo-7a,12a-dihydroxy-5P[7P,acid, was ity of peptides would be the “smuggling in” by coupling to a 12~-SH]cholan-24-oyl)-2-aminoethanesulfonic inhibited by the peptide-bile acid conjugates. In con- natural ligand which is absorbed by a specific transport pathtrast, the parent peptides and amino acids neither had a way, either by receptor-mediated endocytosis as with vitamin significant effect on [SH]taurocholateuptake by ileal B,, (9) or by a carrier mechanism. Our approach makes useof brush-border membrane vesicles nor on photoaffinity the enterohepatic circulation of bile acids involving the small labeling of the ileal bile acid-binding membrane pro- intestine and the liver under physiological conditions (10). This teins. The inhibitory effect of peptide-bile acid conju- organotropism of bile acids is explained by the existence of gates on [3Hltaurocholatetransport and photoaffinity active,Na+-dependentbileacidtransportsystemsin the labeling of the bile acid-binding proteins in rabbitileal plasma membraneof enterocytes and hepatocytes(11-14). The vesicles decreased with increasing chain length of the protein components of the membrane transporters for the upattached peptide radical. By in vivo ileum perfusion in take of bile acids by enterocytes (15-18) and hepatocytes (19, anesthetized rats an intestinal absorption of the bile 20) have been identified with photolabile bile acid derivatives acid conjugate 53744 of the fluorescent oxaprolylpep(21-23). The high capacity of these transport systems recomtide 4-nitrobenzo-2-oxa-1,3-diazol-~-Ala-Phe-5-Opr-Gly (51037) and secretion of the intact compound into bile mends bile acids as a natural shuttling system to improvethe couldbe demonstrated, whereas theparent peptide intestinal absorption of poorly absorbable drugs and to deliver 51037 or its t-butylester 54404 were not absorbed. The drugs selectively to the liver (24), the latter being demonstrated intestinal absorption of53744 showed a similar tem- recently (25-27). perature dependence as [SH]taurocholate absorption EXPERIMENTALPROCEDURES and was inhibited by the presence of taurocholate indicating a carrier-mediated uptake of 53744 via the ileal Materials bile acid transporter. In conclusion, these results indiPhotoaffinity labeling was carried out with the sodium salt of (3,3cate that oligopeptides can be made enterally absorb- azo-7a,l2a-dihydroxy-5~[7P,12P-3Hlcholan-24-oyl~-2-aminoethanesulable by coupling to modified bile acid molecules makingfonic acid (specificradioactivity 5.9 Cilmmol)synthesized as described use of the specific intestinal absorption pathway for bile (21, 22). [3HlTaurocholic acid (specificradioactivity 2.1 Cilmmol) was acids. This finding may be of great importance for the purchased from Du Pont-NEN (NEN Division, Dreieich, Germany).Acdesign and development of orally activepeptide drugs. rylamide, NJV-bismethyleneacrylamide, and marker proteins for the

* Part of this work has been published in preliminary form (55). The costs of publication of this article were defrayedin part by the payment of page charges. This article must therefore be hereby marked “aduertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. $ To whom correspondence should be addressed: SBU Metabolism, Hoechst Aktiengesellschaft, D-65926 Frankfurt am Main, Federal Republic of Germany. Tel.: 0049-69-305-3557;Fax: 0049-69-305-13333.

determination of molecular weights were from Sigma. Cellulose nitrate filters for transport measurements (25-mm diameter, 0.45-pm pore size, ME 25) were from Schleicher & Schull (Dassel, Germany) and scintillators Quickszint 501,361 and Unisolve I from Zinsser Analytic (Frankfurt, Germany). The kits (Merckotest) for the determination of the activity of the markerenzymes aminopeptidase N and y-glutamyltransferase were from Merck (Darmstadt, Germany). Serva Blue R-250 and all other materials for electrophoresis were from Serva (Heidelberg, Germany). All other substances were obtained from the usual commercial sources and were of analytical grade.

10621

10622

Conjugates

Acid

Peptide-Bile

Chemical Synthesis mannitol t o 90 pl of incubation medium containing the radioactively labeled substrate keptat 30 "C. The composition of the incubation meThe fluorescent oxaprolylpeptides S4404, S1037, and their bile acid conjugates S3744 and S2831 were synthesized a s described elsewhere dium for measurements in thepresence of a Na' gradient usually was 10 mM Tris-HCI (pH 7.41, 100 mM NaC1, 100 nm mannitol and in the (28-30). D-Alanine Amino Acid a n d Peptide Deriuatiues-N-Protected D-ala- absence of a Na' gradient, 10 mM Tris-HCI (pH 7.4), 100 nm KC1, 100 nine and (o-alanyl), peptides ( n = 2-4) were synthesized by standard mM mannitol. For measurement of taurocholate uptake, these media contained 50 p~ (0.75 pCi) L3Hltaurocholate and for glucose uptake 19 methods of peptide chemistry (31-33). N-Protection was either by the At desired time points, the transport reacBoc' (tert-butyloxycarbonyl) group or the Z (carbobenzoxy) group. N- p~ (1pci) ~-[U-'~Clglucose. tion was terminatedby the addition of 1 ml of ice-cold stop solution (10 Protected (D-alanyl), (0-prolyl), peptides (n = 3, 6, 8; m = 1, 2) were synthesized by stepwise solid phase peptide synthesis (34) following a n m Tris-Hepes buffer (pH 7.41, 150 nm KCI). The entire content was pipetted onto the middle of a prewashed, prechilled filter kept under orthogonal protecting scheme throughout most of the chain assembly. The acid labile Sasrin resin (35)was chosen as polymeric support and suction with the aid of a vacuum controller. The filter was rinsed immediately with 5 ml of ice-cold stop solution and then solubilized in the base labile Fmoc (9-fluorenylmethoxycarbonyl)group was used for temporary N-protection (36)except for the coupling of the N-terminal scintillator Quickszint 361. The radioactivity remaining on the filter After correcamino acid. In that case the Boc group was used for N-protection in was counted with standard liquid scintillation techniques. tion of medium radioactivity bound tothe filter in the absence of memorder to enhance the solubility of the cleaved peptide at thesubsequent brane vesicles and eventual chemiluminescence, absolute solute uptake reaction with bile acid. Coupling of the amino acids was done with as nanomole/mg protein.All experiments diisopropylcarbodiimide in the presence of 1-hydroxybenzotriazole in was calculated and expressed were performed in triplicate and uptake values are given as mean dimethylformamide. The completeness of the coupling reactions was S.D. monitored by the bromphenol blue method (37). Final cleavage of the N-protected peptide from the resin wasby 1-2% trifluoroacetic acid in Ileal Perfusion dichloromethane. The N-protected peptides were characterized by thinMale Wistar rats were anesthetized with urethane (1.2 gkg) and the layer chromatography, high performance liquid chromatography, and biliary duct was cannulated with polyethylene tubing. Eight cm proxifast atom bombardment-mass spectrometry, mal to the ilea-cecal valve an incision was made into the ileum and a Synthesis of D-Alanyl Peptide-Bile Acid Conjugates-Compounds silicone tubing inserted. A second incision intothe cecum was made and I-XV were synthesized by coupling of the N-protected D-alanine or D-alanyl-peptide derivative to the amino function of 3P-(2-aminoethoxy)- silicone tubing was passed through theileo-cecal valve, fixed in place, 7a,l2a-dihydroxy-5~-cholan-24-oic acid methylester (28-30). In thecase and connected to a peristaltic pump. The loop was filled with buffer I of I-VIII, coupling was done via active ester strategy, utilizing either the(137 nm NaC1,O.g mM CaCI,, 0.51 mM MgCl,, 8.1 mM Na,HPO,, 2.7 mM N-hydroxysuccinimide or 1-hydroxybenzotriazole ester of N-protected KCI, 1.47 mM KH,P04) (pH 7.4), S5% (v/v) ethanol, containing 1 mM at 37"C (12 or o-alanine and of the N-protected D-alanyl peptide, respectively. Com- concentrations of the respective test compounds usually 4 "C in some experiments). The volume of the whole system was 2 ml pounds IX-XV were synthesized using dicyclohexylcarbodiimide and theflow was adjusted to 250 pllmin. For determination of substance o r 2-~'H-benzotriazol-l-yl)-l,1,3,3-tetramethyluronium hexafluoroabsorption, bile was collected over a period of 90 min and samples were phosphate (38) inthepresence of N-hydroxysuccinimide or 1-hydroxybenzotriazole in order toavoid racemization. Deprotection of taken from the perfusion medium and analyzed as thebile samples by the N-protecting group was either done by acidolysis (in thecase of Boc) thin layer chromatography. or hydrogenation (in the case of Z), whereas the methylester group of the Analysis of Bile Samplesby Thin layer Chromatography bile acid was cleaved under basic conditions. The products were purified From the bile samples collected during the ileum perfusion experiby column chromatography and characterized by TLCand 'H NMRspecments, 10-pl aliquots were applied onto high performance TLC plates trometry (100 MHz or 400 MHz). (20 x 10 cm) together with the respective reference standards. Subsequently the chromatograms weredeveloped using thefollowing solvent Animals systems: I, 1-butanollwater/acetic acid, 5/3/2 (v/v/v);11, 1-butanallwater/ Male Wistar rats (Tierzucht Hoechst AG, Kastengrund, Frankfurt acetic acid, 9/2/1 (v/v/v); 111, 1-butanol/water/acetic acid, lO/l/l (v/v/v); am Main, Germany) weighing 300-400 g were maintained on a stanIV, chloroform/methanol, 3/1 (v/v/v). The respective fluorescent comdard diet (AltrominTM) with free access t o water. Food was withdrawn pounds were visualized underUV light and densitometrya t character18 h prior to the ileal perfusion studies. istic ultraviolet absorption at 510 nm using a densitometer CD 50 (DESAGA, Heidelberg, Germany). Quantification wasachieved by caliPreparation of Brush-border Membrane Vesicles bration with definite amounts of the respective reference standards Brush-border membrane vesicles from the ileum of male white New using either the linearor the zig-zag scan mode of the densitometer. Zealand rabbits (weighing 4-5 kg) were prepared by the Mg2' precipitation method (39) as described previously (13, 18,40, 41). The entire Photoaffinity Labeling small intestine was removed and divided into 10 segments of equal Photoaffinity labeling with photoreactive bile acids was performedas length, numbered 1-10, proximal to distal. Segments 8-10 were used described previously (15, 21, 22, 43). Qpically, 15 pl of brush-border for the preparationof ileal brush-border membrane vesicles. The brush- membrane vesicles (150 pg of protein) equilibrated with 10 mM Trisborder membranes were enriched (19 f 4)-fold with regard to aminopepHepes buffer (pH 7.4),300 nm mannitol, were added in the dark 185 to tidase N (EC 3.4.11.2) and (16 2 7)-fold for y-glutamyltransferase (EC pl of 10 mM Tris-Hepes buffer (pH 7.4),100 m NaCI, 100 mM mannitol 2.3.2.2) and freeof contamination by other cell compartments a s shown containing the radiolabeled photoreactive 3,3-azo-derivative of tauroby enzymaticandimmunologicalmethods(18).Immediatelyafter cholic acid and nonradioactively labeled putative inhibitors. After 5 min preparation, the vesicles were stored in liquid nitrogen withoutloss of of preincubation, the suspensions were irradiated for 10 mina t 350 nm transport and enzymatic activityfor at least 4 weeks. The intactness of in a Rayonet Photochemical ReactorRPR-100 (The Southern Ultraviothe vesicles was determined by measuring Na+-dependent D-ghCOSe let Co., Hamden, CT) equipped with RPR 16 3500-A lamps. Afterwards, uptake after 15 s of incubation; usually the overshoot uptake at 15 s was the suspensions were diluted with1 ml of ice-cold buffer (10 mM Trisgreater than20-fold, The enzymatic activitiesof aminopeptidase N and Hepes buffer (pH 7.4), 300 mM mannitol) and centrifugedfor 30 min a t y-glutamyltransferase were determined with Merckotest kits andpro- 48,000 x g. The supernatant was carefully removed and membrane tein was determined according to Bradford (42) using Bio-Rad the assay proteins were precipitated (44). The dried membrane proteins were (Bio-Rad, Munchen, Germany). solubilized in 50pl of 62.5 nm Tris-HC1 buffer (pH 6.8), 2% SDS (w/v), 10% glycerol (v/v), 5% 2-mercaptoethanol (v/v), 0.001 bromphenol blue Dunsport Measurements (w/v), and submitted toSDS-PAGE. Uptake of radiolabeledsubstrates by brush-bordermembrane Gel Electrophoresis vesicles was determined by the membrane filtration method (39) as described previously (13, 17, 18, 40, 41). Typically, the transport reacSDS-PAGE was carried out in vertical slab gels (20 x 17 x 0.15 cm) tion was initiatedby adding 10 pl of the vesicle suspension (50-100 pg using an electrophoresis System LE 2/4 (LKB Pharmacia Biotechnoloof protein) equilibrated with 10mM Tris-Hepes buffer (pH7.4), 300 mM pie, Freiburg, Germany) as described (15, 18, 45). After staining with Serva Blue R 250, the gels were scanned with a densitometer CD 50 of 2 mm thickThe abbreviations used are: Boc, t-butoxycarbonyl; NBD, 4-nitro- (DESAGA) and the individual lanes were cut into slices ness. Eachslice was solubilized with 250 pl of tissue solubilizer Biolute benzo-2-oxa-1,3-diazol; Opr, 5-oxaproline; PAGE, polyacrylamide gel S overnight and after addition of 4 ml of scintillator Quickszint 501 the electrophoresis.

10623

Peptide-Bile Acid Conjugates Ho

MW

MW of peptide radical

536.75

72.09 72.09

-pound I II 111 IV V VI VI1 Vlll IX X XI XI1 Xlll XIV

522.73 670.89

656.06 607.03

593.80

xv

206.22 206.22 143.17 143.17 277.30

741.97 727.94 749.99

277.30

735.96

285.32

776.03 1086.30 107235 1220.54 1214.51

31 1.36 621.71 621.71 763.07 763.07

285.32

0

compound

I

x

S 1037

S4404

s 3744 S2031

I

NH-(CH2),,C0

MW

MW of peptide radical

555.51 61 1.62 1 a31.23

538.50 530.50

1 157.47

664.74

538.50

FIG.1. Peptide-bile acid conjugates.

chain length were covalently attachedtothe 3-position of linker-modified bile acids (Fig. 1)thoroughly considering the structural requirements of bile acids for optimal molecularrecRESULTS AND DISCUSSION ognition by the intestinal transporter(12, 1 8 , 4 6 4 9 ) .Peptides Design of Peptide-Bile Acid Conjugates-For optimal recogcontaining D-amino acids were chosen as model substrates to nition of a bile acid molecule by the active Na+-dependentbile avoid enzymatic hydrolysis of the peptide backbone by brushacid transport systems in the liver and in the small intestine, a border enzymes. In orderto investigate mainly the influence of negative charge in theside chain of the bile acid molecule and at least one hydroxyl group in positions 3,7, or 12 ofthe steroid the peptide chain length on the molecular interaction of the moiety is a prerequesite (12, 18, 46-49). Photoaffinity labeling peptide-bile acid conjugates with the ileal bile acid carrier syswith a bile acid derivative lacking the 12a-oriented hydroxyl tem(s) and to avoid other effects, e.g. effects of functional side acid chains or ordered conformations, no trifunctional amino acid, group, 7,7-azo-3a,12~-dihydroxy-5~[12a-3Hlcholan-24-oic revealed the significance of the l2a-hydroxygroup for optimal but D-alanine as the simplest optically active amino acid was recognition of a bile acid by the ileal bile acid transporter (18) chosen. When synthesizing longer homo-alanyl peptides, it is explaining the low ileal absorption of 12p-hydroxy bile acids necessary to introduce a different amino acid into the peptide known for such as lagodeoxycholic acid (50). In previous studies we could backbone to avoid ordered conformations as they are show that drug-bile acid conjugates with attachment of the homo-oligopeptides. In the case of polyalanyl peptides, a tranrespective drug to the 3-position of the steroid nucleus are sition from random-coil t o p-structure occurs for chain lengths recognized like natural bile acids by the bile acid transport of n = 6 in most organic solvents and in water (51). Furthersystems of the liver (25,27). Consequently, peptides of different more, homo-alanyl peptides containing 10-20 alanyl residues samples were counted for radioactivity. Additionally, radioactivity was detected by fluorography as described elsewhere (18, 45).

Peptide-Bile Acid Conjugates

10624

TABLE I Effect of peptides and peptide-bile acid conjugates on Nu+-dependent uptake of [3Hltaurocholate by brush-border membrane vesicles from rabbit ileum The uptake of [3Hltaurocholate (0.75 pCi, 50 p)into ileal brushborder membrane vesicles was measured for 60 s in the presence of 0, 25,50,100,150,200, and 250 p of the indicated compounds both in the presence and theabsence of an inwardly directed Na' gradient. Inhibition values are expressed as thedifference of uptake in thepresence of Na' minus the absence of Na'.IC, is the concentration of inhibitor where Na+-dependent [3Hltaurocholateuutake was inhibited bv 50%.

PM

72

32 No No

Taurochenodeoxycholate 14 D-Na (D-Ala), (D-Ala), I I1 I11

Iv V

VI VI1 VI11

50

M

II 0

I

I

loo

200

Inhibitor (w)

FIG. 2. Interaction of peptide-bile acid conjugates with Na+dependent [sHltaurocholate uptakeby rabbit ileal brush-border membranevesicles. Ileal brush-border membrane vesicles (10 pl, 50 pg of protein) equilibrated with 10 m~ Ws-Hepes buffer (pH 7.4), 300 m~ mannitol were incubated at 30 "C with 90 pl of 10 m~ TrisHepes buffer (pH 7.4), 100 m~ mannitol, 100 m~ NaCl containing 50 p (0.75 pCi) [3Hltaurocholateand theindicated concentrations of (D-Ala), (O),(D-Ala), (O),or the peptide-bile acid conjugates V, VI, E, X, S1037, S2831, S3744, and taurochenodeoxycholate. Subsequently, [3H]taurocholate uptake was measured for 1min by a rapid membrane filtration method. Upper panel: 0, (o-Ala),; 0 (D-Ala),; B, compound V; 0, compound VI; compound X , 0 , compound M;X , taurochenodeoxycholate. Lower panel: 0, NBD-P-Ala-Phe; 0,S1037; B, S2831; 0,,33744; X , taurochenodeoxycholate.

X XI XI1 XI11

m xv

S 1037

S 4404 s 3744 S 2831

inhibition inhibition No inhibition 48 220 ND 143 69 84 270 126

ND

76 123 >300 62 >300 280 ND 235 ND 238 ND ND 270 No inhibition No inhibition 89 ND 240

108 ND" 225 225 164 219 212 250 250

179 >300 >300 >300 >300 >300 >300 ND

ND ND ND ND

ND

250 ND

ND

ND, IC value not determined in the concentration range used.

gates and behaving as taurocholate during enterohepaticcirculation (21). Photoaffinity labeling of the bile acid-binding proteins of M , 93,000 and 14,000 which have been identified as the essentialprotein components of the Na+hileacid co-transporter from rabbit ileum (17,18), was inhibited by the peptideare highly helical (52, 53). The helical structure leads to poor bile acid conjugates, whereas the free peptides did not show a coupling yields at the stepwise formation of homo-alanyl pep- significant inhibitory effect on photoaffinity labeling of these tides using solid phase peptide synthesis (52).To avoid these bile acid-binding proteins (Fig. 3). The labeling of the integral difficulties D-proline was introduced at position 2 (XI), 4 and 8 bile acid-binding membrane proteinof M , 87,000 thought tobe (XI-XV)of the peptides for D-proline is known to disrupt pep- involved in membranebile acid transport by a facilitatedtranstide a-helices (54). port system (18)was also inhibited by peptide-bile acid conjuInteraction of Peptide-BileAcidConjugates with the Ileal gates. These transport and photoaffinity labeling studies demNu+-dependentBile Acid nansport System-Molecular recog- onstrate a specific interaction of the peptide-bile acid nition of the peptide-bile acid conjugates with theileal Na'hile conjugates with the ileal bile acid carrier system(s1. In Vivo Ileal Absorption of Peptide-Bile Acid Conjugates-In acid co-transport system was investigated by measuring the effect of these compounds and of the corresponding free oli- order to investigate the intestinal absorption of peptide-bile gopeptides on the uptake of [3Hltaurocholate by brush-border acid conjugates in a living animal, peptide-bile acid analogues membrane vesicles prepared from rabbit ileum. Fig. 2 shows containing additionallya fluorescent reporter group(NBD, Fig. that peptide-bile acid conjugates ledto a concentration-depend- 1) were studied in a n in vivo ileum perfusion model. After ent inhibitionof [3H]taurocholate uptake, whereas thepeptides cannulation of the common bile duct, therespective compounds alone had no significant inhibitory effect on bile acid uptake. were instilled as a 1mM solution into a closed loop ileal segment Attachment of peptide radicals greater than4 amino acid resi- and recirculated at a flow rate of 0.25 mumin. Bile was colwere taken from the a significant dropin theaffinity of the peptide- lected over a period of 90 min and samples dues resulted in bile acid conjugates to the ileal bile acid transport system as collected bile and the perfusion medium for analysis by thin indicated by an increaseof the IC,,, IC,,, or IC,, values (Table layer chromatography. Since the intestinal absorption is the rate-limiting step during enterohepatic circulation of bile acids, I). To evaluate whether thesynthesized peptide-bile acid conju- the appearanceof bile acids in bile can be used to determine the gates arerecognized by the ileal Na+-dependentbile acid trans- intestinal absorption of bile acid derivatives. As a model compound for these investigations we have choport system, their interaction with the protein components of this transport system wasinvestigated by photoaffinity label- sen S3744, where the NBD-modified tetrapeptide p-Ala-PheOpr-Gly was linked to the 6-atom spacer modified cholic acid ingusing (3,3-azo-7a,12a-dihydroxy-5~[7~,12~-3Hlcholan-24oyl)-2-aminoethanesulfonic acid,a photolabile taurocholate derivative 3~-(5-aminopentoxy)-7a,l2a-dihydroxy-5~-cholan24-oic acid. To detect an intestinal absorption of NBD-p-Alaanalogue (21) being modified by the photoreactive diazirino (azi) group in the 3-position as in the peptide-bile acid conju- Phe-Opr-Gly in S3744 by making use of the ileal bile acid

+

Peptide-Bile Acid Conjugates 1

66

--

43

-

116 97

3

E:

2

5

4

5

6

7

8

9

1

0

1

10625

1

,.

36,

D

L

i -29 24 20

14

-

FIG.3. P h o t o a f f h i t y l a b e l i n g of rabbit ileal brush-border membranevesicles with (3,3-azo-7a,12a-dihydroxy-5P[7P,12PSHlcholan-24-oyl)-2-aminoethanesulfonic acid in the presence of peptides and peptide-bileacid conjugates. Ileal brush-border membrane vesicles ( 15 PI, 150 pg of protein) were mixed for 5 min in the dark with 185 pl of 10 m r Tris-Hepes buffer(pH 7.4), 100mM mannitol, 100 m r NaCl containing 1.27 1 - 1 ~(1.5 pCi) (3,3-azo-7a,12a-dihydroxy5~17~,12~-~'Hlcholan-24-oyl~-2-aminoethanesulfonic acid without or with 250 PM concentrations of the indicated compounds. Subsequently, the vesicle suspension was irradiated a t 350nm for 10min. m e r washing thevesicles, membrane proteins were separated by SDS-PAGE (12% gels) and radioactivity was detected by fluorography. Lanes: I , control; 2, D-Ala; 3, (n-Ala),; 4 , (D-Ala).,;5, compound 111; 6, compound W, 7, compound I; 8, compound 11; 9, compound VII; 10, compound VIII; 11, compound V.

transport system distinct from an absorption by other mechanisms, the intestinal absorption of the parent peptide S1037 and a lipophilic prodrug thereof, the t-butylester S4404, was also investigated in an identical experimental setup. Neither the parent peptide S1037 (Fig. 4A) nor its t-butylester S4404 (Fig. 4B) appeared in bile and furthermore,also no fluorescent fragments of these compounds possibly generated by luminal, membrane, or intracellular hydrolysis could be detected in bile. These findings clearly indicatethat peptides S1037 and S4404 are not absorbed by the rabbit ileum. In contrast, after instillation ofthe bile acid conjugate S3744 into anileal segment, the intact peptide-bile acid conjugate was secreted into bile with a secretion profile similar to a natural bile acid (Fig. 4C) with a secretion maximum after 12-16 min compared with 8-12 min for taurocholate. In addition, a metaboliteMI was secreted into bile with a secretion profile parallel to the parent compound S3744. Thin layer chromotographic analysis of bile and perfusion medium probes in different solvent systems proved the identity of MI detected in bile after ileal perfusion with S3744 with MI formed intraluminally from S3744, S1037, and S4404. The exact structure of MI is not known but presumably is a derivative of NBD-P-Ala-Phe-Opr. Control experiments with M I isolated from the perfusion medium revealed that MI did not inhibit [3Hltaurocholate uptakeinto ilealbrush-border membrane vesicles. Furthermore, since the metabolite M, is formed intraluminally from compounds S3744, S1037, and S4404 and since no M, appears in bile after ileal perfusion with peptides S1037 and S4404, it can be excluded that the amounts of MI detected inbile after ileal perfusion with S3744 arise from an intestinal absorption of MI. The MI detected in bile after ileal perfusion from S3744 is formed from S3744 during liver passage as could be shown by liver perfusion experiments after bolus injection of S3744 into a peripheral mesenteric vein (25). Within a collection period of 80 min, around 1.1%of the applied amount of S3744 was secreted into bile in intact form and a further 1.15%as metabolite MI giving a resorption rate of >2.25% for S3744 within the collection period. Analysis of the

3

4

@

:2 '85 2 0 3 0 4 0 6 0 8 0

B L 8 12 16 20 30 w) 60 80

Time (min)

FIG.4. In situ ileal perfusion with fluorescent peptides and peptide-bile acid conjugates. The fluorescent peptides S1037 and S4404 and the peptide-bile acid conjugates S3744 and S2831 (1 mhl), dissolved in buffer I, were instilled intoan ileal segment (10cm length) of anesthetized rats and recirculateda t a flow rate of 0.25 mVmin after cannulation of the common bile duct. Bile samples and perfusion medium werecollected after the indicated time points and analyzed by thin layer chromatography (10-pl aliquots) in different solvent systems (I, 1-butanovacetic acid/water,5/2/3 (v/v/v); 11, 1-butanovacetic acidwater, 9/1/2 (v/v/v); 111, 1-butanovacetic acidlwater, lO/l/l (v/v/v); IV, chlorofodmethanol, 3/1 (v/v), andvisualization of the fluorescent NBD-labeled substrates under U V light. (The chromatograms shown were developed in solvent system 111.) The figures represent the time points (in min)of bile collection after instillation of the compound into an isolated ileal segment.B,control bile; F, solvent front;S , start zone; G , fluorescent compound present in bile unrelated to peptide andpeptide-bile acid conjugates; MI, metabolite of S3744, S4404, and S1037. A, peptide S1037;B , peptide S4404; C , peptide-bile acid conjugate S3744; D,peptide-bile acid conjugate S2831; E , time dependence of the biliary secretion of S3744 after in situ ileal perfusion. Thin layer chromatograms of bile samples were analyzed by densitometry a t 510 nm and quantification was achieved by calibration with definite amounts of S3744.0, peptide-bile acid conjugate S3744; 0, metabolite M Iof 53744.

perfusion medium revealed that S3744 was intraluminally rapidly degraded to metabolite MI; after 20 min around 30% of S3744 remained as intactbile acid conjugate S3744 within the intestinal lumen. Since the metabolite MI is not transferred from the intestinal lumen into bile to a significant extent, an intestinal absorption ratearound 7% can be estimated for S3744 in the chosen experimental setup. Since the ileal uptake of bile acids is an active, Na+-dependent transportprocess, uptake of ligands by this transport system should greatly depend on temperature. Therefore, in vivo ileal perfusion experiments were carried out where the ileal segment was placed into an external bath kept a t 12 or 4 "C and

10626

Peptide-Bile Acid Conjugates

10 IO

20

30

40

50

60

70

eo

20

30

40

SO

60

T m ie 0

FIG.6. Inhibition of ileal absorption of 53744 by taurocholate. T m ie 0 A solution of 1 m S3744 dissolved in buffer I, with or without 2 m FIG.5. Temperature dependence of ileal absorption of [%]tautaurocholate, was instilled into an ileal segment of anesthetized rats rocholate and 53'744. [3HlTaurocholate(3 m)or S3744 (1m)dis- and recirculated at a flow rate of 0.25 mumin. Bile was collectedat the solved in buffer I were instilled into an ileal segment (10 cm length) of indicated time points and analyzed for S3744 by thin layer chromotoganesthetized rats and recirculated at a flow rate of 0.25 mumin after raphy. 0 , secretion profile of S3744.0,secretion profile of S3744 in the cannulation of the common bile duct. Bile was collectedat the indicated presence of taurocholate. time points and analyzed either by liquid scintillation counting ([3Hltaurocholate)or fluorescence measurement after thin layer chro- acid transport system (17, 18) by (3,3-azo-7a,12a-dihydroxymotography. Upperpanel, secretion profile of 3 m [3Hltaurocholateat 5~[7~,12~-3H]cholan-24-oyl)-2-aminoethanesulfonic acid was 12 "C (0-40 min) and after raising the temperature to 37 "C (40-80 min). Lower panel, secretion profile of 1 m S3744 at 12 "C (040min) concentration dependent inhibitedby the peptide-bile acid conjugates. In contrast, the corresponding parent peptides neither and after raising the temperature to 37 "C (40-80 mid.

inhibited taurocholate transport nor photoaffinity labeling of closed-loop perfusion was performed for 40 min with bile col- the transportproteins. The affinity of the peptide-bile acid conlection after the indicated time periods. Then, theileal loop was jugates to theileal bile acid carrier system decreased with inof the model peptide. By in vivo ileal pertransferred back to the abdominal cavity (temperature 37 "C) creasing chain length fusion experiments it could be shown that linearoligopeptides and subsequentlybile was collected at the indicated time points with modification of the N terminus by the fluorophore NBD (in after loop transfer. At 12 or 4 "C a continuous secretion of to the molecular volume of the [3H]taurocholate into bile was observed. After raising the tem- size approximately corresponding or tryptophan) became intestinally abamino acids tyrosine perature to 37 "C, immediately an increaseof biliary secretion of [3Hltaurocholate occurred with a typical secretion maximum sorbed after coupling to the 3p-position of a 5-atom linkerpeptide itselfwhich was not caused by the onset of active ileal bile acid absorption (Fig. 5, modified bile acid in contrast to the transported. A strong temperature dependence of uptake of the upper panel). After instillation of S3744 into an ileal segment at 12 or 4 "C, no significant amounts of the parentcompound or peptide-bile acid conjugate S3744 and its secretion into bile as its metabolite M, appeared in bile; however, after raising the well as a n inhibition of uptake by taurocholate strongly argues temperature to37 "C, ,33744 was secreted into bile showing the for a carrier-mediated ileal uptake via the Na+hile acid co5, lower transport system. After replacement of the 4-atom spacer (NHtypicalsecretionmaximumliketaurocholate(Fig. bile acid conjugate of NBD-0panel). A carrier-mediated ileal uptake of S3744 was further CH,-CH,-CO) of p-alanine in the Ala-Phe-Opr-Gly (S3744) by a 13-atomlinkerin12confirmed by inhibition of biliary secretion of 53744 by the aminododecanoic acid (NH-(CH,),,-CO) with conservation of all simultaneous presence of taurocholate in the ileal perfusion other structural elements, no significant transport of the cormedium (Fig. 6).In order to obtain first insights into the structure-activity relationships of peptide-bile acid conjugates for responding bile acid conjugateS2831 from the intestinal lumen intestinal absorption, the p-alanine spacer groupbetween the into bile could be detected. Sincep-alanine wasreplaced by the NBD-receptor group and the oxaprolylpeptide in S3744 was lipophilic amino fatty acid 12-aminododecanoic acid, it is posreplaced by the lipophilic 12-aminododecanoic acid 62831). In sible that theincreased lipophilicity of the peptide backbone in contrast to S3744, no significant amounts of S2831 or fluores- S2831 leads to a sticking of the molecule to biological membranes. InS3744, an 18-atom backbone made of a tetrapeptide cent fragmentsthereof could be detected in bile (Fig. 40)after and anaminopentyl-spacer groupattached to the 3p-position of instillation of this compound into the ileal lumen; the comcholic acid was inserted between the NBD-reporter group and pound remained unchanged in the perfusion medium. the bile acid molecule leading toa significant intestinal absorption of the peptide-bile acid conjugate. Since the space-filling CONCLUSIONS size of the (NBD-p-Ala-Phe-Opr-Gly-NH-(CH,),O)-moiety atIn order to evaluate the active, Na+-dependentileal bile acid tached to cholic acid corresponds approximatelyto thatof a linuptake system asa means for the intestinal absorption of pepear heptapeptide,it seems probable that modified bile acid moltides, a series of small linearmodel peptides up toa chain length ecules can be used as a "hook-up" system to improve the intesof 10 amino acids were covalently coupled to the 3-position of tinal absorption of therapeutically useful oligopeptides via the peptidyl-3p-(w-aminoalkoxy)modified bile acid yielding active Na+-dependentbile acid transport system in theileum. 7a,l2a-dihydroxy-5/3-cholan-24-oic acids. These compounds These findings may be a first step toovercome the obstacle of were able to interact with the ileal Na'hile acid co-transport nonabsorbable biologically active peptides and be of great imsystem as was shown by a concentration-dependent inhibition portance for the development of orally active peptide drugs. of Na+-dependent [3Hltaurocholate uptake into brush-border membrane vesicles from rabbit ileum. Photoafinity labeling of Acknowledgments-We thank Susanne Winkler and Silke Elsesser the proteincomponents of M,93,000 and 14,000 of the ileal bile for excellent secretarial assistance.

Conjugates

Acid

Peptide-Bile

REFERENCES 1. Holmes, C. P., Adams, C. L., Fodor, S. P. A., and Yu-Yang, P. (1993) in Perspec-

2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13.

tiues in Medicinal Chemistry (Testa, B., Kyburz, E., Fuhrer,W., and Geiger, R., eds) pp. 489-500, Verlag Helvetica Chimica Acta, Basel Smith, P. L., Wall, D.A.,Gochoco,C. H., and Wilson, G. (1992) Adu. Drug Deliuery Reu. 8, 253-290 Lee, V. H. L. (1988) CRC Crit. Reu. Thel: Drug Carrier Syst. 5, 69-97 Matthews, D. M. (1991) in Protein Absorption: Development and Present State ofthe Subject, Wiley-Liss, New York Saffran, M., Kumar, G . S., Savariar, C., Burnham, J. C., Williams, F., and Neckers, D. C. (1986) Science 233, 1081-1084 Lee, V. H. L., andyamamoto, A. (1990)Adu. Drug. Delivery Reu. 4, 171-207 Ziv, E., Kidron, M., Berry, E., and Baron, H. (1981) Life Sci. 29,803409 Fujii, S., Yokoyama, T., Ikegaya, K., Sato, F., and Yokoo, N. (1985) J. Pharm. Pharmacol. 37,545-549 Russell-Jones, G. J., and Aizpurna, H.J. (19881Proc. Znt. Symp. Contl:Release Biouct. Mater. 15, 142-143 Vlahcevic, 2. R., Heuman, D.M., and Hylemon, P. B (1990) in Heptatology (Zakim, D., and Boyer, T. D., eds) pp. 341477, W. B. Saunders, Philadelphia, PA Wilson, F. A. (1981) Am. J . Physiol. 241, 6 8 3 4 9 2 Lack, L. (1979) Enuiron. Health Perspect. 33, 79-90 Burckhardt, G., Kramer, W., Kurz, G . , and Wilson, F. A. (1983) J . Biol. Chem.

258,3618-3622 14. Schwarz, L. R., Burr, R., Schwenk, M., Pfaff, E., and Greim, H.(1975) Eul: J . Biochem. 55, 6 1 7 4 2 3 15. Kramer, W., Burckhardt, G., Wilson, F. A,, and Kurz, G . (1983)J . Biol. Chem. 258,3623-3627 16. Lin, M. C., Weinberg, S. L., Kramer, W., Burckhardt, G., and Wilson, F.A. (1988) J. Membl: Biol. 106, 1-11 17. Kramer, W., Nicol, S:B., Girbig, F., Gutjahr, U.,Kowalewski, S., and Fasold, H. (1992) Biochim. Biophys. Acta 1111,93-102 18. Kramer, W., Girhig, F., Gutjahr, U., Kowalewski, S., Jouvenal, K., Muller, G . , Tripier, D., and Wess, G. (1993) J. Biol. Chem. 268, 18035-18046 19. Kramer, W., Bickel, U., Buscher, H . 2 , Gerok, W., and Kurz, G. (1980) HoppeSeyler’s Z.Physiol. Chem. 361, 1307 20. Kramer, W., Bickel, U., Buscher, H:P., Gerok, W., and Kurz, G. (1982) Eul: J . Biochem. 129, 13-24 21. Kramer, W., and Schneider, S . (1988) J . Lipid Res. 30, 1281-1288 22. Kramer, W., and Kurz, G. (1983) J. Lipid Res. 24,910-923 23. Kramer, W. (1981) Die Zdentifizierung Gallensaure-bindenderPolypeptide

durch Photoaffinitatsmarkierung.Synthese und Anwendung Photolabiler Deriuate der Gallensauren, pp. 1-316, Inaugural-Dissertation, Universitat Freiburg 24. Ho, N. F. H. (1987) Ann. N. Y. Acad. Sci. 907, 315-329 25. Kramer, W., Wess, G . , Schubert, G., Bickel, M., Girbig, F, Gutjahr, U., Kowalewski, S., Baringhaus, K.-H., Enhsen, A., Glombik, H., Mullner, S., Neckermann, G . , Schulz, s.,and Petzinger, E. (1992) J. Biol. Chem. 267, 1859818604 26. Stephan, Z. F., Yurachek, E. C., Sharif, R., Waswary, J . M., Steele, R. E., and Howes, C. (1992) Biochem. Pharmacol. 43, 1969-1974 27. Kramer, W., Wess, G., Schubert, G . , Bickel, M., H o h a n n , A., Baringhaus,

K.-H., Enhsen, A., Glombik, H., Miillner, S., Neckermann, G., Schulz, S., and Petzinger,E. (1993) in Bile Acids andthe Hepatobiliary System (Paumgartner, G., Stiehl,A,, and Gerok, W., eds) pp. 161-176, Kluwer Academic Publishers, Dordrecht

10627 28. Wess, G . , Kramer, W., Bartmann, W., Enhsen, A,, Glombik, H., Mullner, S., Bock, K., Dries, A., Kleine, H., and Schmitt,W. (1992) Tetrahedron Lett.33, 195-198 29. Wess, G., Kramer,W., Enhsen, A., Glombik, H., Baringhaus, K.-H., Bock, K., Kleine, H., and Schmitt, W. (1993) Tetrahedron Lett. 34,817-818 30. Wess, G., Kramer, W., Schubert, G., Enhsen, A,, Baringhaus, K.-H.,Glombik,

H., Mullner, S., Bock, K., Kleine, H., John, M., Neckermann, G., and Hoffmann, A. (1993) Tetrahedron Lett. 34,819-822 31. Houben-Weyl (1974) MethodenderOrganischenChemie, (Muller, E., ed) Fourth Ed., Vol. XV/1 and XV/2, Thieme-Verlag, Stuttgart 32. Bergmann, M., and Zervas, L., (1932) Bel: Dtsch. Chem. Ges. 65, 1192-1201 33. Brand, E., Erlanger, B. F., and Sachs, H., (1952) J . Am. Chem. Soc. 74,18491851 34. Memfield, R.B. (1963) J. Am. Chem. Soe. 85,2149-2154 35. Mergler, M., Tanner, R., Gosteli, J., and Grogg, P. (1988) Tetrahedron Lett.29, 4005-4008 36. Atherton, E., and Sheppard, R. C. (1987) in The Peptides, Analysis Biology (Gross, E., and Meienhofer, J., edsj Vol. 9, pp. 1-38, Academic Press, New

York 37. Krchnak, V., Vagner, J., Safar, P., and Lehl, M. (1988) Collect. Czech. Chem. Commun. 53, 1542-1548 38. Knorr, R., Trzeciak, A., Bannwarth, W., and Gillessen, D. (1989) Tetrahedron Lett. 30, 1927-1930 39. Hopfer, U., Nelson, K., Perotto, J., and Isselbacher, K. J. (1973) J. Biol. Chem. 248.25-32 40. Kramer, W. (1987) Biochim. Biophys. Acta 905,65-74 41. Kramer, W., Girbig, F., Leipe, I., and Petzoldt, E. (1988) Biochem. Pharmacol. 37,2427-2435 42. Bradford, M. M. (1976) Anal. Biochem. 72,248-254 43. Kramer, W., Buscher, H.-P., Gerok, W., and Kurz, G . (1979) Eul: J. Biochem. 102, 1-9 44. Wessel, D., and Fliigge, U. J. (1984) Anal. Biochem. 138, 141-143 45. Kramer, W., Girbig, E , Gutjahr, U., and Leipe, I. (1990) J . Chromatogl: 521, 199-210 46. Anwer, M. S., OMaille, E. R. L., Hofmann, A. F., DiPietro, R. A., and Micheloff, E. (1985) Am. J. Physiol. 249, G479-G488 47. Aldini, R., Roda, A,, Morselli Labate, A. M., Cappelleri, G., Roda, E., and Barbara, L. (1982) J . Lipid Res. 23, 1167-1173 48. Hardison, W. G . M., Heasley, V. L., and Shellhamer,D. F. (1991)Hepatology 13, 68-72 49. Lack, L., and Weiner, I. M. (1967) Biochim. Biophys. Acta 135, 1065-1068 50. Schmassmann, A., Angellotti, M. A,, Clerici, C., Hofmann, A. E , Ton-Nu, H.-T.,

Schteingart, C. D., Marcus, S. N., Hagey, L. R., Rossi, S. S., and Aigner, A. (1990) Gastroenterology 99, 1092-1104 51. Mutter, M., Pillai, V N. R., Anzinger H., Bayer, E., and Toniolo, C. (1981)

Peptides (1980)Proceedings ofthe Sixteenth European Peptide Symposium (Brunfeldt, K., ed) pp. 66-65, Scriptor, Copenhagen, Denmark 52. Memfield, R. B., Singer J., and Chait, B.T. (1988)Anal. Biochem. 174, 399414 53. Marqusee S., Robbins, V. H., and Baldwin, R. L. (1989) Proc. Natl. Acad. Sci. U. S. A. 86, 528G5290 54. Chou, P. Y.,and Fasman G . D. (1974) Biochemistry 13, 222-245 55. Kramer, W., Wess, G . , Baringhaus, K.-H., Boger, G . , Enhsen, A,, Falk, E.,

Friedrich, M., Glombik, H., Hoffmann, A,, Neckermann, G . , Pittius, C., and Urmann, M. (19931 Hepatology 18, 299